Masaharu Yoshioka

Supporting conceptual design based on the function-behavior-state modeler

The relative significance of conceptual design to basic design or detail design is widely recognized, due to its influential roles in determining the products fundamental features and development costs. Although there are some general methodologies dealing with functions in design, virtually no commercial CAD systems can support functional design, in particular so-called synthetic phase of design. Supporting the synthetic phase of conceptual design is one of the crucial issues of CAD systems with function modeling capabilities. In this paper, we propose a computer tool called a Function-Behavior-State (FBS) Modeler to support functional design not only in the analytical phase but also in the synthetic phase. To do so, the functional decomposition knowledge and physical features in the knowledge base of the modeler, and a subsystem Qualitative Process Abduction System (QPAS) play crucial roles. Modeling scheme of function in relation with behavior and structure and design process for conceptual design in the FBS Modeler are described. The advantages of the FBS Modeler are demonstrated by presenting two examples; namely, an experiment in which designers used this tool and the design of functionally redundant machines, which is a new design methodology for highly reliable machines, as its application.

Physical concept ontology for the knowledge intensive engineering framework

Knowledge intensive engineering aims at flexible applications of a variety of product life cycle knowledge, such as design, manufacturing, operations, maintenance, and recycling. Many engineering domain theories are organized and embedded within CAD and CAE tools and engineering activities can be formalized as modeling operations to them. Since most of domain theories deal with the physical world and can be associated with physical concepts, a physical concept ontology can form a common ontology to integrate engineering models that are formed based on domain theories. This paper reports a physical ontology-based support system for knowledge intensive engineering called Knowledge Intensive Engineering Framework (KIEF) to integrate multiple engineering models and to allow more flexible use of them. First, the paper describes the physical ontology as the core of KIEF and an ontology-based reasoning system, called a pluggable metamodel mechanism, to integrate and maintain relationships among these models. The pluggable metamodel mechanism uses a metamodel that represents the designers mental model about a design object as a concept network model. The designer builds and decomposes a functional hierarchy from functional specifications with an FBS (Function-Behavior-State) modeler. He/She then maps the functional hierarchy into a metamodel using physical features that are building blocks for conceptual design. Then, the pluggable metamodel mechanism enriches the information contained in the metamodel by using causal dependency knowledge about the physical world and by building and analyzing various engineering models. We demonstrate the power of KIEF by illustrating a design case performed on KIEF.

The CHEMDNER corpus of chemicals and drugs and its annotation principles

The automatic extraction of chemical information from text requires the recognition of chemical entity mentions as one of its key steps. When developing supervised named entity recognition (NER) systems, the availability of a large, manually annotated text corpus is desirable. Furthermore, large corpora permit the robust evaluation and comparison of different approaches that detect chemicals in documents. We present the CHEMDNER corpus, a collection of 10,000 PubMed abstracts that contain a total of 84,355 chemical entity mentions labeled manually by expert chemistry literature curators, following annotation guidelines specifically defined for this task. The abstracts of the CHEMDNER corpus were selected to be representative for all major chemical disciplines. Each of the chemical entity mentions was manually labeled according to its structure-associated chemical entity mention (SACEM) class: abbreviation, family, formula, identifier, multiple, systematic and trivial. The difficulty and consistency of tagging chemicals in text was measured using an agreement study between annotators, obtaining a percentage agreement of 91. For a subset of the CHEMDNER corpus (the test set of 3,000 abstracts) we provide not only the Gold Standard manual annotations, but also mentions automatically detected by the 26 teams that participated in the BioCreative IV CHEMDNER chemical mention recognition task. In addition, we release the CHEMDNER silver standard corpus of automatically extracted mentions from 17,000 randomly selected PubMed abstracts. A version of the CHEMDNER corpus in the BioC format has been generated as well. We propose a standard for required minimum information about entity annotations for the construction of domain specific corpora on chemical and drug entities. The CHEMDNER corpus and annotation guidelines are available at: http://www.biocreative.org/resources/biocreative-iv/chemdner-corpus/

Representation of Design Object Based on the Functional Evolution Process Model

One of the crucial issues for developing computer aided conceptual design system is representation of functions which represent designers’ intention. Representing functions is also crucial not only for representing design objects but also for describing conceptual design processes, in which designers operate mainly functional concepts. Namely, function is a key concept to integrate object modeling and process modeling in design. In this paper, first we extend the FBS (Function-Behavior-State) diagram, which we have already proposed, by introducing three additional concepts for representing a function; namely, function body that represents designers’ intention directly, function modifier that qualifies a function body, and objective entity on which the function body occurs. This extended FBS diagram, called FBS/m (modifier) diagram, enables us to represent designers’ intention more precisely than the original FBS diagram. Then, we propose an FEP (Functional Evolution Process) model to represent design processes. In the FEP model, the FBS model of a design object is evolved through three steps, i.e., functional actualization, functional evaluation and functional operation. Functional actualization depicts a process to obtain physical descriptions from functional description. Functional evaluation is a process to measure realizability of functions of the design object. Functional operation is a process to operate functions to improve the design. Based on the FEP model, we analyze an actual design process, and show that the FEP model is suitable for representing designers’ intention along with design processes.

Knowledge systematization for a knowledge intensive engineering framework

This paper proposes knowledge intensive engineering that is a new way of engineering activities in various product life cycle stages flexibly conducted with more knowledge to create more added value. Knowledge representation and modeling issues are discussed and a cooperative multiple intelligent agent architecture based on multiple ontology is proposed for building a Knowledge Intensive Engineering Framework (KIEF). KIEF can be used as a knowledge intensive CAD for knowledge intensive design of knowledge intensive machines. This demonstrates the power and usefulness of knowledge intensive engineering. It is also discussed that to achieve knowledge intensive engineering, systematization of knowledge is an essential process to allow intelligent agents to share accumulated knowledge.

ABDUCTION FOR CREATIVE DESIGN

While abduction is considered crucial for design in general, this paper focuses on the role of abduction to integrate knowledge assuming that creative design can come from innovative combination of existing knowledge. Based on Schurz’s classification of abductive reasoning, the paper identifies that abduction for integrating theories can be performed by a special type of abduction called second order existential abduction. The paper then analyzes refrigerator design cases to understand how knowledge is used and shows that abduction is indeed central to design. It also discusses that knowledge structure is a key concept in abduction for integration.Copyright

Design for upgradable products considering future uncertainty

Conventional products tend to be thrown away because of functional obsoleteness before reaching the end of the physical life of the products. This is one of the main causes of the mass disposal problem. In order to solve this problem, the paper proposes a design methodology for upgradable products. In particular, the paper discusses a method dealing with uncertainty caused by long-term planning in the design methodology for upgradability. The proposed method deals with uncertainty as ranges of parameter values, and derives a design solution that realizes required upgrades, while adapting to estimated uncertainty. This method makes the products robust and tolerant against uncertainty, and therefore makes design for upgradability more feasible.

Applying a Burst Model to Detect Bursty Topics in a Topic Model

This paper focuses on two types of modeling of information flow in news stream, namely, burst analysis and topic modeling. First, when one wants to detect a kind of topics that are paid much more attention than usual, it is usually necessary for him/her to carefully watch every article in news stream at every moment. In such a situation, it is well known in the field of time series analysis that Kleinberg’s modeling of bursts is quite effective in detecting burst of keywords. Second, topic models such as LDA (latent Dirichlet allocation) are also quite effective in estimating distribution of topics over a document collection such as articles in news stream. However, Kleinberg’s modeling of bursts is usually applied only to bursts of keywords but not to those of topics. Considering this fact, we propose how to apply Kleinberg’s modeling of bursts to topics estimated by a topic model such as LDA and DTM (dynamic topic model).

A knowledge operation model of synthesis

This chapter describes an attempt to formalise and model synthesis theoretically. It aims at establishing unified understanding of synthesis, beginning with an analysis-oriented thought process model and a synthesis-oriented thought-process model. These models are given logical interpretations to be performed on a multiple model-based reasoning framework. We then analyse design activities and show that design, including synthesis and analysis, is largely a knowledge-based activity. This results in a knowledge operation model that is decomposed into logical operations and modelling operations. The core of synthesis is considered to be abduction, and, within our framework, abduction is realised as model-based abduction. We outline an algorithm for model-based abduction. The knowledge operation model of synthesis was tested against an actual design case from which a reference model was built. The knowledge operation model was also implemented on the multiple model-based reasoning framework and the reference model was performed on it to perform the verification of the knowledge operation model.

IR interface for contrasting multiple news sites

In order to utilize news articles from multiple news sites, it is better to understand the characteristics of each news site. In this paper, a concept of contrast set mining is applied for analyzing the characteristic difference between each news site and all others. The News Site Contrast (NSContrast) system is also proposed based on this mining technique. This system is applied to a news article database constructed from multiple news sites from different countries in order to evaluate its effectiveness.